--- title: "VAST Vector Search" slug: "dev-vector-search" updated: 2026-11-05T22:26:58Z published: 2026-11-05T22:26:58Z canonical: "docs.vastdata.com/dev-vector-search" --- > ## Documentation Index > Fetch the complete documentation index at: https://docs.vastdata.com/llms.txt > Use this file to discover all available pages before exploring further. # VAST Vector Search VAST Vector Search is nearest-neighbour search over embeddings stored in VAST DataBase tables. New to Vector Search? Read the Overview, run the Quickstart, then read The vector index to learn the rules a search must follow. Before you create a real table, work through Design a vector table and Create a vector index, then follow the guide for your task, such as Ingest vectors or Query design. ## Overview and concepts Where to begin: the overview and quickstart, then the concepts and architecture the rest of the section relies on, and the release notes. - **[Overview](/develop/docs/dev-vector-search-overview)**: Nearest-neighbour search over vector columns in VAST DataBase tables, queried with SQL through the VAST Query Engine, for application developers. - **[Quickstart](/develop/docs/dev-vector-search-quickstart)**: Create a vector-indexed table, insert rows and find nearest neighbours with SQL, then repeat the steps with the Python SDK. - **[The vector index](/develop/docs/dev-vector-search-concepts-index)**: Objects, lifecycle and distance values of a vector-indexed table, and the rules a search must follow to use the index. - **[Vector Search architecture](/develop/docs/dev-vector-search-concepts-architecture)**: Where a vector index lives on the cluster, how it is built in the background, and what happens during a search. - **[Release notes](/develop/docs/dev-vector-search-release-notes)**: What VAST Cluster 5.5 supports for vector search, what each 5.5 release added, and the known limitations. ## Guides Step-by-step instructions for building with Vector Search. Start with Design a vector table and Create a vector index, then Ingest vectors and Query design; the other guides cover filters, quantization, measuring recall, migration and RAG retrieval. - **[Create a vector index](/develop/docs/dev-vector-search-guides-create-an-index)**: Create a vector-indexed table with SQL or the Python SDK, choose its metric and quantization, and avoid rejected definitions. - **[Ingest vectors](/develop/docs/dev-vector-search-guides-ingest)**: Load embeddings with SQL, the Python SDK, ADBC bulk ingest or Parquet import, and change a row's embedding afterwards. - **[Quantization](/develop/docs/dev-vector-search-guides-quantization)**: The two quantization settings of a vector-indexed table, the combinations VAST Cluster 5.5 accepts, and how to trade accuracy for speed. - **[Query design](/develop/docs/dev-vector-search-guides-query-design)**: Write searches in the shape the index recognises, match distance functions to metrics, and rewrite queries that don't use the index. - **[Filters](/develop/docs/dev-vector-search-guides-filters)**: Combine nearest-neighbour search with WHERE conditions, plan filter columns before creating the table, and filter results by user permissions. - **[Design a vector table](/develop/docs/dev-vector-search-guides-table-design)**: Plan the index, metric, quantization and columns of a vector table before CREATE TABLE, since most choices can't be undone. - **[Measure recall and latency](/develop/docs/dev-vector-search-guides-developer-observability)**: Time vector queries from the application, measure recall, tune the probe settings against it, and give support a query ID. - **[Build RAG retrieval with permissions](/develop/docs/dev-vector-search-guides-rag-retrieval)**: Chunk, embed, load and retrieve documents for RAG, returning only chunks from documents the requesting user may read. - **[Move vectors into an indexed table](/develop/docs/dev-vector-search-guides-migrate)**: Copy vectors into an indexed table from an unindexed table, pgvector, Pinecone-style hosted services, or Spark and Trino pipelines. ## Integrations Client details that are specific to vector search. Start with ADBC for Vector Search, the main client; read Python SDK for Vector Search if your application uses vastdb. - **[ADBC for Vector Search](/develop/docs/dev-vector-search-integrations-adbc)**: Connect to the VAST Query Engine with adbc_driver_vastdb, set the options that matter for searches, and bulk-ingest vectors. - **[Python SDK for Vector Search](/develop/docs/dev-vector-search-integrations-python-sdk)**: Create vector-indexed tables, insert and import rows, read the index definition and search with table.vector_search() in the vastdb SDK. - **[LangChain](/develop/docs/dev-vector-search-integrations-langchain)**: Use a VAST DataBase table as a LangChain vector store with langchain-vastdb: add documents, search, filter, retrieve for RAG, and see the SQL it runs. ## Reference Exact syntax, settings and limits to look up. Start with the VAST Vector Search cheat sheet; go to the SQL reference, Search session settings or Limits of VAST Vector Search when you need the full detail. - **[SQL reference for VAST Vector Search](/develop/docs/dev-vector-search-reference-sql)**: Vector column types, the CREATE TABLE ... VECTOR CLUSTER BY grammar, distance functions and the errors the Query Engine returns. - **[Search session settings](/develop/docs/dev-vector-search-reference-configuration)**: Names, formats and defaults of the ADBC session settings that control how much of the vector index a search reads. - **[Limits of VAST Vector Search](/develop/docs/dev-vector-search-reference-limits)**: Index, table and search rules for vector-indexed tables in VAST Cluster 5.5, most of them fixed at table creation. - **[VAST Vector Search cheat sheet](/develop/docs/dev-vector-search-reference-cheat-sheet)**: One-page summary of creating, searching and connecting to vector-indexed tables, with quantization pairs and links to the details. ## Operations What to arrange with your platform team, and how to run vector search in production. Start with Setup to get Query Engine access; read Security, Performance and sizing, and CI/CD and automation as you move toward production. - **[Setup](/develop/docs/dev-vector-search-operations-setup)**: What to request from your platform team for Query Engine access, a copyable request, and a check that searches work. - **[Security](/develop/docs/dev-vector-search-operations-security)**: Which actions, row filters and column rules apply when searching vectors through the VAST Query Engine, and which don't. - **[Performance and sizing](/develop/docs/dev-vector-search-operations-performance)**: Check index state, reduce what a search reads, spread load and estimate the space a vector table uses on the cluster. - **[CI/CD and automation](/develop/docs/dev-vector-search-operations-ci-cd)**: Create vector-indexed tables from a pipeline with the VMS REST API, vastpy, or Terraform with the Python SDK. ## Troubleshooting Help when a search, load or connection goes wrong. If you have an error message, search the Error index first; otherwise start from the symptom tables in Troubleshooting, or the FAQ for common questions. - **[Troubleshooting](/develop/docs/dev-vector-search-troubleshooting-index)**: Symptoms, likely causes and first fixes for connection, permission, table creation, search and data-change problems with vector search. - **[Error index](/develop/docs/dev-vector-search-troubleshooting-errors)**: Messages from creating, loading, changing and searching vector-indexed tables, each with its cause and fix, searchable by message text. - **[FAQ](/develop/docs/dev-vector-search-troubleshooting-faq)**: Short answers on vector search basics, the index, searching and data protection, each linking to the page with details.